The Science of Everything Podcast - Episode 67: An Overview of Language Part 2

Episode Date: November 23, 2014

Concluding the brief exploration of language begun last time, this episode examines semantics, including the distinction between sense and reference and different types of utterances, pragmatics, including conversational maxims and implicatures, and sociolinguistics, including prestige dialects and speech acts. Recommended pre-listening is Episode 66: An Overview of Language Part 1.

You're listening to The Science of Everything podcast, episode 67, an overview of language part two. And I'm your host, James Fodor. So in the last episode, we looked at language, and I gave an overview of the different subfields of linguistics. And I talked about the first few of those, so phonetics, phonology, and morphology. In this episode, I'm going to continue that discussion and look at some of the other areas that we didn't get to, syntax, semantics, pragmatics, and sociolinguistics. So, let's jump right in. So let's move on from syntax to semantics, which is my personal favorite. subfield of linguistics, and I'm not sure if linguists would agree with this, but I would

tend to say that it's the one we know the least about, it's the one that's sort of least well developed. This is how you actually get meaning out of all of those phonemes and morphemes and all the syntax and put all that together, but how does the meaning actually come out of it? What is it that distinguishes colorless green ideas sleep furiously from that? What does it distinguishes that sentence from I had a good sleep last night? Both of them are grammatical, and both of them use, you know, the same collection of morphemes and phonemes and things like that. They're both in the same language, but one makes sense and the other one doesn't make sense. One has meaning, the other one doesn't have meaning.

What is it that distinguishes those two? What's going on there? How is it that we get meaning out of abstract arbitrary units of sound or squiggles on a paper at all? I mean, when I say tree, that's just an arbitrary collection of sounds. What does it have to do with this thing that has bark and leaves and grows? How do you get the idea of tree in your mind when I say this string of symbols, or when I write down this string of symbols, in case of written language, or when I say this string of sounds in the case of spoken language,

what is the relationship between these arbitrary abstract symbols and the actual meaning behind them? And that's a very hard question to answer, which has connections to all sorts of ideas in psychology and, of course, linguistics, artificial intelligence, and formal theories of grammar and logic, and it gets very, very complicated. So one issue here is so-called underspecification.

meanings are not complete without a lot of context and background knowledge. So if I say that something is red, for example, what that actually means is going to vary a lot, depending on what I'm talking about. Whether it's a red wine, or in which case it's quite dark red, or red hair is actually sort of more like orange and copper sort of color, red soil, which isn't really red, but a sort of dark, richy idea. Red skin, that, again, is a sort of a particular hue of skin color, which isn't really red either, or a red rose, which is a brighter red. So that even basic things like colour words will be understood in very different ways depending on the context that I'm speaking in. And this will lead us later into pragmatics. But it's not just in terms of conversational context. It's in terms of you have to understand what a red wine is to know what the word red means in the context of red wine.

If you're thinking of red in terms of red hair when I say red wine, then you're not getting the right meaning. That's not what it actually means. It refers to a different color. And it's not even just a color. Red wine also refers to a whole bunch of things related to that. concept. But similarly, if you're thinking red hair, when I'm talking about red skin, again, there's a difference in meaning, and you're just going to misunderstand what I'm saying. You can't convey that meaning. You can't get that meaning out of what I said just by looking at

the words. The word red is the same in all of those cases. What's different is the context, what words that are near it, background knowledge and assumptions of the speakers hold in common, etc., etc. Working out this unsuspicification of meaning is actually a very tricky thing to do. There's also sort of even the simpler case of where we have the same word that means many, many different things. Stream, for example, can mean all sorts of different things. Get is, I think, the word that has the most different meanings in English, or one of them. Set is another example that means just all sorts of different things,

and how do we go from the same sound to the meaning out of it? Well, it's from context and background assumptions and all sorts of things that allow us to infer meaning from it, and that's really tricky. So if we're talking to someone who has very different assumptions, very different knowledge of the world or very different experiences, it's going to be difficult to communicate with them, because when we use words, they go into different connotations, different meanings, different associations.

So one influential and sort of more traditional approach in semantics, which is still useful, and I don't want to say it's obsolete, but it has significant limitations, I think, is called truth-conditional semantics. And the basic approach is to say that the meaning of a sentence is reducible to its truth conditions, or the meaning of a sentence is the set of conditions in which it is true. It is evaluated to have the value of truth.

So saying something like snow is white, the meaning of that is simply the set of conditions in which it is true, the set of conditions in which it is true to say that snow is white. One problem with this is that it has the implication, at least in its traditional formulation, that all necessary truths mean the same thing, have precisely the same meaning. So according to this, saying that 2 plus 2 equals 4, which is necessarily true, means the same thing as saying an apple is an apple, because that's also necessarily true. the set of conditions, the set of conditions in which they are true is the same, the set of all conditions, because they're always true. They're true in all conditions. But that means they, according to this account, that they mean the same thing, which seems nonsense, because 2 plus 0 equaling 4 has nothing to do with an apple being an apple, or bachelor being unmarried might be another example. So this seems to have significant limitations, although it's still very useful when you're looking at language in terms of formal logic and formal theories of grammar, because being able to apply truth functions to things and work out the truth values of complex sentences and things, is is a very useful approach, but it doesn't seem to quite capture the richness in which we use language. Another limitation of this relates to a related idea called prototype theory, which is a reaction to the idea that we establish definitions for words, that a word means one thing, and that's its definition,

that you have necessary and sufficient conditions. Something is a cat, if it has four legs and fur and a tail and a bunch of other things, and it must have those things, otherwise it's not a cat, and if it has those things, it is a cat, and that's as far as it is. According to this traditional idea of necessary and sufficient conditions, we can define exactly what we mean by a table and a chair and a book and a country and beauty and morality and everything. We can define them all precisely in terms of necessary and sufficient conditions, and that's what we mean when we say any of these words. We're just referring to its sort of list of defining conditions. Now, a moment's reflection, I think, clearly indicates that this just doesn't work.

Wittgenstein argued sort of persuasively about this, but many others have too. that almost all concepts that we use are fuzzy, that there are no necessary insufficient conditions. You just try to find a table, for example. It's essentially impossible to provide necessary and sufficient conditions as to what a table is or a chair. I mean, a chair is something like something that you sit in, but, I mean, what about chairs in a doll's house

that no one ever sits in, or chairs that are purely ornamental, and I'm not designed to be sat in? One can go on and on. Any particular criteria that you give, you can always find an exception to it. What if I have a cat? Well, what if I shave a cat? Is it still a cat? Well, yes. What if a cat loses a leg? Is it still a cat? Yes.

So it doesn't have to have four legs. It doesn't have to have fur. If I cut its tail off, it's still a cat. You might say, well, what about if it has a certain genetic set up? Well, if I did gene therapy on the cat, would it still be a cat? Well, probably would be, unless I did enough gene therapy to make it something else. But also, it seems that we can know what cats are without knowing anything about genes. And so it seems odd to define a cat in terms of its genetics, because that's not what we're really talking about when we talk about cats. So any of these sorts of things, Nesser and sufficient conditions just don't work, or at least for most concepts that we use. And so instead, an alternative approach to characterising the meaning is to talk about prototypes.

The idea is that there are, and again, we talked about this in the knowledge representation episode, so I won't dwell on it too much, but the idea is that there are some cases which are clear, paradigmatic cases of a category, like a clear example of a chair. Maybe for you, that has a wooden back and it has four legs and user for sitting at a table. Well, maybe that's the paradigmatic case of a chair. And, you know, a sort of a lounge chair or something like that or a stool there, they're sort of chairs, but they're not exactly as chair-like as the paradigmatic case of the chair. Paranomatic case of a cat is, I don't know, a cabby cat or something like that, with it's four legs and furmen, it's all stuff like that. And if I start talking about, well, if it loses a leg, well, it's a little bit less like a cat,

because normally I think about a cat having four legs, not three. If I talk about tigers, well, they're sort of cats, but kind of not cats. So the idea of prototype theory is that we have sort of a, sort of a, fuzzy membership. Some things are central and paradigm cases, and then other things are progressively less and less like that. So there's no necessary and sufficient conditions. The meaning is fuzzy in this way. That's an important idea in semantics. Another important idea in semantics is the difference or distinction between sense and reference. This is an idea that was developed by a philosopher called Frege. Now, the difference between sense and reference is what is a little

bit hard to understand, but let me try to explain it. They both relate to what a term means, but means in a different sense. So the sense of a term or a word is what exactly is being referred to, what specifically is being said. The reference is the sort of object, the reference, the thing in the world that's being picked out, the object that the expression refers to. So sense is sometimes described as a mode of presentation. It's the way that we're presenting the underlying thing. That's the sense, the way it's presented. The underlying thing is the reference. Okay, so some examples here.

So, a classic example is Venus. Venus is the reference, it's the planet Venus, but there are multiple senses ways in which we can refer to it, the morning star and the evening star. Those are the same thing. They're both the planet Venus. But they're different senses, if I say the morning star,

it's not quite the same as when I say the evening star. I'm presenting it in a different way. Another example would be for people who have multiple different names. So Mark Twain and Samuel Clemens, they are the same person. They pick out the same person in the world. Their reference is the same, but they have a different sense. Because if I talk about Mark Twain, that's a different sense to Samuel Clemens,

because they're different senses, they're different modes of presentation. So sense is also referred to as the cognitive significance, because if I say a book is by Mark Twain, that has more cognitive significance, at least to a lot of people than Samuel Clemens. But they actually refer to the same thing, their reference is the same, but their sense is different. So this is a useful concept for understanding meaning, because we might have the same, when we say,

well, what does that mean? Are we talking about the sense or the reference? And that can, understanding the difference can help us to disambiguate certain cases where we talk about meaning. One final point that I just want to make in regards to semantics is the function of sentence, or the purpose of articulating a particular sentence, the purpose of engaging in communication.

So there are often described as four basic functions of sentences, four basic purposes to communication. Four reasons why we might say something. These are declarative, interrogative, exclamative, and imperative. So declarative is sort of the most common type of sentence. This is just when we make a statement where we assert something. The cat is on the mat. That's a declarative sentence. I just said something.

I am recording a podcast. That is a declarative sentence. Most sentences usually a declarative. That's sort of the default. Although some people have sort of questioned that as to sort of reifying declarative sentences being the most important type of thing which decided. But anyway, I think usually we do make declarative sentences. Interrogatives are questions.

So what are you doing? I'm not stating something. I'm asking a question. I'm asking for information from you. So that's an interrogative. An exclamative is an exclamation. So if I bang myself on the furniture, I might say, ouch, that's an exclamation.

Or if I'm really happy, I might say, you pee! that's an exclamation. I'm not asking for any information, nor am I stating anything, I'm just really expressing myself in some way. I might be conveying some meaning as well. Like if I say, ouch, or ow, maybe I'm conveying a meaning that I want you to attend to me or that I need help, but it's not quite the same. Indeed, if I yell help, that's an exclamation. That conveys meaning, but I'm not really, I'm not making a declaration. So it's not that declamations don't convey meaning, but they conveyed in a sort of a different way to a declarative statement. And imperative, the last type, is an order.

It's an injunction. I'm telling you to do something. Wait for me here. That's an injunction. I'm not asking a question, nor am I just exclaiming. I'm telling you to wait here. It's a command. Now, it's interesting that the same word can function as, or the same sentence,

can function as declarative, interrogative, exclamative, or imperative, depending on the context. and intonation and use of certain auxiliaries and particles and other things like that. So an example that I think is interesting is the word no. You can use the word no just by itself in any of these senses. So it can be declarative. If someone asks you a question, you respond, you say no. Is Berlin the capital of France? No.

Really what I'm saying there is Berlin is not the capital of France. I'm making a declaration. But of course, I'm sort of, because of context, I'm shortening it down to just saying no. But that's really a declaration. I'm asserting a fact. Is it true that such and such? No. I'm saying it's false that such and such. That's a declaration. On the other hand, I can ask a question. I could say, no. So maybe you just told me that Berlin is not the capital of France, but I thought it was for some reason. And so I could say, no. I'm asking a question, basically. I'm saying, isn't it the capital France? They're not serving spaghetti in the cafeteria today. No? Because they usually serve spaghetti of the cafeteria today. So I say no in a sort of confused way. That's really asking a question. I'm asking for clarification. Well, aren't they? Or I thought this was the case. I'm not declaring something I'm asking. It's an interrogation. I'm essentially asking for clarification or asking you to respond.

Well is another example. Well, you know, we're asking for someone to expound on something. As an exclamative, well, that's very easy. I can just say, no. Maybe if I'm very upset about something, I say no! That's clearly an exclamation. I'm not asking or saying something, I'm exclaiming. An imperative, again, very clear for the case of no,

if the child is about to touch the al-stove, I can point and yell, no. That's an imperative. I'm telling them not to. The key point here is that the same word can function in very different ways depending on how it's used as a declarative, interrogative, exclamative or imperative, and it's quite interesting to make a distinction in terms of understanding the difference in meaning between those, the meaning that's conveyed. Okay, moving on from semantics now to pragmatics, getting to sort of the quite abstract levels of language here. Pragmatics is the subfield of linguistics which studies how context contributes to meaning, and so we touched on that earlier by looking at the, you know, interrogatives and exclamatives and so on, because which of those is meant can often be determined by context.

But semantics is different to pragmatics because semantics generally focuses on the meaning that's internal to the language, that's encoded in the particular utterance or text that's being spoken or said or written. On the other hand, pragmatics looks at external factors. So who says it, what they are doing when they say it, the circumstance surrounding them, all of that sort of stuff, the pre-existing knowledge of those involved, the intent of the speaker. Those are all the sort of external to the speech itself or to the text, to the written language itself. So there's a sort of a difference there.

Pragmatics looks at meaning, but meaning from a sort of an external sense, outside the actual language use itself, the context surrounding the language use, whereas semantics looks meaning from an internal point of view within the language itself. There's obviously a sort of a fuzzy line between them, and they interact with each other, but I think you can draw a distinction between them. So let's give an example.

If I say, you have a green light. Now, that sentence is, you know, well-formed, and it seems to convey some meaning, but what is actually meant? What am I actually saying when I say, you have a green light? Well, maybe I'm a stage director, and I'm telling the actors on stage

that they are being illuminated with green ambient lighting. That might be relevant to them for the performance in some way. It could mean that I'm waiting at the traffic lights, and I'm not paying attention, and someone in the cast telling me that I've got a green light. So what they're actually telling me is that I should start driving now.

That's a very different meaning to the ambient lighting for the actors. It might mean, if I say you have a green light, it means you should go ahead with something. You have permission to do something. It could mean that your body has a green glow to it. That would be kind of an odd usage, but, you know, theoretically, it could mean that. It could mean literally that you possess a light bulb that's tinted green. You have a green light. I don't have any green lights at the party.

No, you do have a green light because I gave you one last week. So very different senses, very different meaning conveyed in all these cases from exactly the same sentence. Just looking at the sentence you have a green light by itself won't tell you what was meant by those, because you need to know the context. Am I on stage or sitting in a car or are we in a business meeting or are we talking in the context of some party where we're talking about the decorations? Is this some weird medical condition where the hue of my body would be relevant? So the context and what's meant and the background knowledge of the speakers is going to be crucial to understanding which of these meanings is relevant. And that's where pragmatics comes in. Pragmatics is what studies this. Another interesting example that I like is the sentence Sherlock saw the man with binoculars.

We could indeed talk about this in terms of syntax because there's different ways of passing the sentence. Is the man with binoculars the object? Is it the person who Sherlock saw? is it Sherlock saw the man with binoculars, in which case the man with binoculars is an object, or is the man the object, and with binoculars is the manner in which Sherlock saw the man? So is it Sherlock saw the man with binoculars, or is it Sherlock saw the man with binoculars? Which of those is true will depend on the context. If I knew that Sherlock was holding binoculars, if that was told to me earlier, or the binoculars were specifically mentioned in some way, or that I knew that the man was very far away,

and so he had to be using binoculars or something like that, then I would know that it meant one sense. On the other hand, if I knew that Sherlock was looking for a man holding binoculars for some reason, then I would know it was meant in a different sense. So a context, again, is going to determine the meaning in that instance. Now, a very important principle in pragmatics is what's called the cooperative principle,

which also, and is related to this idea of conversational maxims, these are supposed to be sort of rules of thumb that we all adhere to in normal conversation and that we must adhere to, otherwise we can't make sense of what other people are saying. And when people violate these axioms, then we have potential confusions. So these maxims, there are four of them that are sort of traditionally put forward, and I'll just explain them briefly. The maxim of quality essentially says that you shouldn't say things that are false

or for which you lack adequate evidence. The maximum of quantity says that your contribution to the conversation should be as informative as necessary, but no more. So don't say more than you need to, but don't say less than you need to. The maxim of relevance essentially just says be relevant and don't say things that aren't important to the purpose at hand. And the maxim of manner says that you should avoid obscure expressions. You should be clear and orderly in the way you present your ideas.

So when these maxims are violated, it can lead to confusion, it can lead to misunderstandings, people even becoming upset. On the other hand, they can also be deliberately violated for purposes of sarcasm, irony. comedians often deliberately violate the axioms in order to produce a comedic effect. Violation of the axioms can also be used to convey meaning in various interesting ways. So let me just give an example, though, of violating these maxims. So someone who tells a bold-faced lie is clearly violating the maxim of quality, that you're not supposed to say things that you believe are false.

This is not a moral statement we're making here. It's not saying you shouldn't lie. It's just saying that if we're to understand each other, if we're to engage in meaningful conversation, there needs to be a presupposition that we generally say things that, are true, because if we just lied perpetually and constantly, we would never be able to make any sense of what was being said, because we'd have no idea if it was true or not. So we need to have this basic pre-supposition that we're not telling falsehoods. And similarly, that we don't just assert things for which we lack evidence. You might know some people who habitually just say things that are not necessarily

false, but they just don't have evidence for. They just sort of say stuff, and like, how can you possibly know that? You might know that you have to be careful with these people, because normally you expect people to not say things unless they have some bases for it. Maybe it's not a great basis, but at least some basis for it. But then there are some people who just seem to say stuff, and you have to be careful with them because you know not to take what they say quite so seriously. The maximum relevance is a particularly easy one to understand, because if I ask someone a question and they respond by saying something completely different, what they said might be perfectly meaningful. It might make perfect sense and be well-formed syntactically and all that,

but it just doesn't have anything to do with what I asked. Or maybe it just doesn't have anything to do with what I was talking about before. And so that leads to confusion. Well, hang on, why did you say that? Well, are they trying to get at something or they're trying to make some point? It's just weird. It's confusing when people just randomly change the subject like that. It's because it's violating the maximum of relevance. It's also a bit weird or off-putting or confusing. When people either speak too long for a context or too short for a given context, this violates the maximum of quantity. So if I give a very long, detailed explanation of something to someone, and then you just say, okay, that's not a sufficient response. They need to

give some more feedback. Is the part that they didn't understand, or did they understand it all fine, or have they heard it before? Something more. Similarly, if you spend a long time, I don't know, narrating a story to someone or explaining how you feel or all of these sorts of things, it's not sufficient just to say a very brief one or two-word response. It's violating the maximum of quantity. You need to say a little bit more. On the other hand, if someone's, if we ask someone a question about, I don't know, some question about what they purchased at the supermarket, and they go into this long spiel about the exact colour of all the different eggs that they considered and the number of people in the aisle and how the lighting was and all that's other

things. That's violating the maxim of relevance, because we just don't care about that stuff. It's not important, and it's also violating the maximum of quantity because they're giving us far too much information that's irrelevant. So it's confusing, we're wondering, what's the point are they trying to get at something, and it's irritating. So violating these maxims can make conversation quite difficult, but they can, can also be used, as I said, for comedic or sarcastic effect. And, as I'm now moved to now, they can also be used to convey meaning in various interesting ways, or we can use them to infer meaning in various interesting ways.

So this leads to the notion of something called conversational implicature, which is a technical term which refers to things that are implied by what we say, even if they're not entailed by them. So it's not directly said in what we said, or it's not directly entailed by it. It's not logically derived from what we said, but it is certainly implied by what we said. So this is a very important way that meaning is conveyed. It's conveyed by implicature. But implicature is very complicated and hard to understand because it depends on all sorts of things.

So an example that I stole from Wikipedia, if I say that Mary had a baby and got married, that sort of suggests that she had a baby and then got married. But it doesn't actually say that because it strictly the sentence is true, even if Mary had the baby after she got married. But it's an implication of that, of what I said, because of various background assumptions. As another example, I can violate the maximum of relevance if I'm wanting to avoid hurting someone's feelings.

Or maybe if I'm wanting to be sarcastic. So if someone asks me, how did you like the talk? I could say, well, it was in English. I understood the words. Now, that seems to be kind of irrelevant. Like, well, why are you saying that? But the implication is, well, I couldn't say any. anything else. Maybe in order to avoid violating the maximum of quality, that is, in order

to avoid lying, I need to say something else. So I have to say something that's less relevant. Well, it was in English. Another example might be if someone asks, or do you like this dress, do I look good in this dress? Maybe I say something like, I like the colour. Now, that's potentially violating the maximum of quantity or maybe of relevance, because maybe I should say more or I should say something that's more directly relevant about how the person looks, not the color of the dress, but maybe I'm concerned about violating the maximum of quality. I don't want to lie and say something that's false.

I don't want to say they look good in the dress when I actually think that they don't, and so I say something else. So there's, and this is how we can draw implications from what people say. They avoid violating a maximum of conversation, perhaps by violating another one, or just by saying something in particular,

and we can infer from that, so the reason they said this thing, The reason they said that they like the colour of the dress is because they couldn't say that they think I look really good in the dress. And so I can infer from that that they don't like the dress. They don't like how the dress looks on me. Even though they didn't say that, I'm inferring that from what they said because of the conversational maxims. An example about the maximum relevance. If someone asks me, is there somewhere where I can get gas, get petrol, and I say, oh, yeah, there's a station around the corner,

that would be, strictly speaking, completely irrelevant if the station was closed or didn't. sell gas or petrol. So by saying that, the person infers from what I said that, yes, there is somewhere where I can get petrol or gas, and it's nearby, and it's open, and I'll be able to purchase gas there and so on. But I didn't actually say any of those things. They're inferring it from the fact that what I said would have been, strictly speaking, irrelevant if those things weren't true. If the place was closed, then it would be irrelevant that it was there, and you should have told me that it was closed. So those are relating to the maxims of quantity and relevance. So this is another example of how we can infer meaning and how we do infer meaning based on appealing to these conversational maxims.

Another example, related to the maxim of quantity is the idea that we expect people to make the strongest claim that they can, and is also that's relevant to the case, to the situation. So this is called scalar implicature. So if I say something like, I ate some of the pie, now that's strictly speaking true regardless of how much of the pie I ate. I could have ate a tiny piece or I could have eaten the whole thing. But we don't expect people to say, I ate some of the pie if, in fact, they ate the whole pie, because that's not the strongest claim they could have made. They should have said, I ate the whole pie, that's a stronger claim, and they could have made it. So if someone says, I ate some of the pie, whereas we assume from that, we infer from that that they only ate some of it, and they didn't eat all of it.

Because if they did eat all of it, they would have said that they ate all of it, because that's a stronger claim. Again, we generally find it strange if people don't make the strongest claims that they can. It's confusing. Maybe if they say something like, there is at least one. skyscraper in New York City, we think, well, okay, yeah, that's strictly speaking true, but why would you say that? You can make a much stronger claim. There are lots of skyscrapers in New York City. We would normally expect people to say that, and not there is at least one. Even if it's sort of strictly speaking irrelevant to the particularities, maybe we actually

only care if there's one. But normally you would just make the strongest claim possible. You wouldn't make this weird weaker claim when you could make the stronger claim. And conversely, of course, if you don't make the stronger claim, that implies that you couldn't have made it because we expect you to do so. Like, for example, did you like the movie? Well, there are parts of it that I liked. Well, the fact that I like parts of it is not inconsistent with me liking the whole thing, but because I expect you to make the strongest claim possible and say, I liked it, if indeed you did like it, the fact that you didn't say that, and the fact that I assume that you're following the conversational maxims means that I must infer, ah, you couldn't make that claim, because if you did, it would violate the maximum of being truthful. So from that I infer, well, you mustn't have liked the whole movie, and so you're just saying you like parts of it. None of that follows logically from what was said, but it's implied based on the deriving from the conversational maxims. There's another interesting concepts in pragmatics, which I just want to mention briefly,

which is the idea of indexical language. An indexical language refers to utterances or communications whose meaning is intrinsically related to the when and the who and the where that they are made. So if I say something like, I went there, the meaning of that depends on who sense it. If I say it, it might be true, but if you say it might be false. If I say, I live here, well, the meaning of that is very different to when I say I live here to when you say I live here, because you'll be a different person saying it in regard to a different place. I live at my house, that's indexical, it's relative to me and my house, but you live at your house.

But you could say I live at my house. When you say I live at my house, that means something different to when I say I live at my house. So it's words like this that refer to the me or also time, so now, or this or also place. of this place. Those are indexical. And the meaning of those in trying to work out how that fits exactly and how we infer the meaning from those is actually quite interesting. And that's part of pragmatics because it all relates to the context in which the language is used. Who is saying it and in what context is they're saying. And from that we can try and infer what they actually mean, what they're saying. The final area of linguistics and language that I wanted to talk about

is what I'm calling sociolinguistics. This is the study of how society and cultural norms and expectations and so on, interact with language and shape the way language is used and understood. So, for example, a sociolinguist might be interested in studying how social attitudes determine the types of vernacular language that is used, or the types of informal language, or the type of register that are used when talking to someone, or different dialects in what circumstances it might be appropriate to use certain slang or a certain way of speaking and so on, and the factors that shape this. An interesting concept here is that of the speech community, which describes a particular distinct group of people who use language in a particular unique way.

So a speech community doesn't have to be a country. It doesn't even have to be a very large group of people. It could be a subculture or it could be, I don't know, a political party that if they have their own sort of in-group language and particular phrases and in-group jokes and things like that, it could even be a family who form their own speech community. And of course, you could be part of more than one speech community. Friendship groups may also form a speech community to an extent. Often they'll have their own slang or jargon or particular presuppositions, ways

of using language, ways of pronouncing words, focuses and priorities, etc., etc. So sociolinguistics will be interested in seeing how different speech communities use language and how the situation of the speech community affects how they use language and how other people perceive them. Now, very important in socio-linguistics is this concept of prestige, that certain habits of speech are assigned positive or negative value when applied to the speaker depending on social context. So this could be at the level of particular sounds, or at the high level of the type of words that people use and when they use them or the way they use grammar.

So, you know, if I use the word ain't, for example, that has a, well, in many contexts at least, a negative connotation. It sounds like I'm not a well-educated or not as careful in my language, but I don't know as much of what I'm talking about. A lot of slang terms have negative connotations, although the connotation applied to something, the prestige associated with it, might differ between speech communities. If I use a particular slang term, that might indeed me,

a particular speech community, whereas if I was in a different context, it might be seen as a negative thing. In general, though, there are certain types of language, words, and particular pronunciations and accents, which are considered to be prestigious in particular contexts. So language is bound up in culture in this way. So a very classic example of this is a so-called received pronunciation, which is, which some people refer to as a posh British accent. This is the way that, it's also referred to as BBC English, because BBC presenters, at least more traditionally used, and I think they're moving a little bit away from it now. But it's the way the Queen speaks, David Attenborough. It's that form of sort of sophisticated English accent.

Now, even the fact that we use a word like sophisticated tells us that this is a prestige accent here, because there's no way in which merely pronouncing words in a particular way can be sophisticated. There's nothing intrinsic to the words that makes them more sophisticated or that makes them convey meaning more clearly. it's simply that it's associated with these positive characteristics, and maybe it is the case that people who speak in this way, on average, have a higher education or something like that. That might be an empirical question, but it's sort of beside the point. The point is that there is a connotation and there's a positive attribution to the person who speaks in this particular way. There are other sorts of dialects or of accents or ways of speaking or words which have negative associations. And often this can be deeply political, which is partly why I'm hesitant to give very specific examples. but I think that everyone will know what I mean when I say that you can often tell something about a person,

or at least you think you can tell something about a person based on how they speak, the words they use and the way they pronounce them and all of those sorts of things. Whether they're accurate or not, well, is interesting, but kind of a different question. The point is that we make these inferences about people. That's very much of interest to socio-linguistics and relates to prestige, so that can affect how you're seen as a political candidate or in a job interview or as a student talking to a professor or anything like this. People will judge you and will make inferences about your character and competence based on how you speak,

even though, strictly speaking, that's kind of irrelevant to whatever the consideration is that they're interested in. Now, also related to this idea of prestige is the concept of a dialect. What is the difference between a language and a dialect? I suppose the core idea is that dialects are different versions of the same language, whereas different languages are sort of different things, different languages. but there is no actual neat distinction between a dialect and a language. In reality, what we observe in the real world is that people who live, certainly this is true more traditionally, it's a little bit different now that we have sort of a global world, but basically

the closer together people live, the easier it is for them to understand each other and the more similarly they speak and the more words in common they have. And as you move further and further apart, their speeches get more and more and different until you reach the point where they're not even speaking the same language anymore. And intermediate to that are points of different dialects and sub-dialects and accents and all these sorts of other things that are various gradations. And, of course, you can group languages into families of languages, which are different languages, but actually quite similar to each other. So Italian and Spanish are quite similar in many ways, and then there's still different languages. Within each of those, there'll be different dialect of them.

And German and English are quite different as languages, but they're also quite similar as languages because they are related to each other historically. They're much more similar to each other than they are, say, to Chinese. So there really is no meaningful distinction or clearly meaningful distinction between a dialect and a language. A sort of a famous aphorism about this which I really love is simply that a language is a dialect with an army and a navy. The idea there is that there isn't actually any difference between a language and a dialect. The only real difference, there isn't any linguistic difference, I should say. The difference is political about how much prestige or power a particular dialect is able to acquire, and thereby they sort of establish their own particular way of speaking as the correct ways.

speaking or as definitive of a language. And historically, you can see this, for example, many of the European courts attempted to establish their own particular dialect or way of speaking as the universal standard throughout the nation. So France in particular did this, with the French kings attempting to establish the Parisian style of French as being the true French that everyone throughout France should speak. And this was done in China as well, with the Mandarin Chinese as being sort of the correct version of Chinese, which everyone was supposed to speak.

And it becomes very political, and so hence, a language is a dialect with an army and a navy. There's one final concept that I want to talk about in social linguistics, although sometimes it's also considered to be part of pragmatics. They certainly overlap. And this is the idea of speech acts. Now, the idea of a speech act is that it is something that you do, some acts that you perform in the very act of speaking itself or of communicating itself. So it's different from just saying something.

if I say, I'm going to go over there. If I say I'm going to go over there, that's not the same as going over there. They're different things. So that's not a speech act. I have to actually go over there. Or I'm going to cook dinner. I cooked a dinner. Those aren't speech acts.

But some things, the act is committed in the speech itself. And there are different types of these, so I'll go through them. These are called illocutionary acts. You're doing something by using language itself. Not that there's some consequence of using the language. You said something and then someone gets upset. It's the language use itself is the act. So there are assertive speech acts where the speech acts commits the speaker to the truth of some express proposition in a very sort of rigid or literal or straightforward way.

Not just sort of I said that something was true, but I am committing myself to the truth of these things. So reciting a religious creed would be an example of that. Or maybe reading a corporate mission statement or something like that. That would be an assertive speech act. You're not just saying something. You're committing yourself to the truth of it. You're binding yourself to it in some sort of real way, some public way. So directives are another type of speech act.

These are ones in which we cause the hearer to take a particular action. So commands are like this, where when we give someone a command, we're not just sort of suggesting that they do something, or telling them to do something, where in the act of saying it, we are trying to change their behavior. A military command might be an example of this. So I gave you an order, when I give an order as a military commander, that's not just sort of something I say.

Giving the order is an act itself, and it has real sort of meaning and power as an act, and that would be a directive. A request would be another example. If I formally request your help for something, or even informally, but maybe it helps the thing about it in a formal sense, that is an act. It's an act of requesting assistance. It's doing something.

Commissives are speech acts which commit you to some future action. So this is a promise or an oath. Again, making a promise is not just sort of saying something. It's doing something. You are committing yourself. You're binding yourself in a public way, or maybe even a legally binding way. Expressives express attitudes and emotions.

So that's like congratulations and thanks. If I say, thank you very much for this, again, I'm not, I mean, I'm expressing myself, but it's more than that. I am doing something. I am doing, making an action by making a thank you or making an excuse or congratulating someone. It's often I think it's helpful to think about these in a formal context, but it doesn't have to be formal. It's very much that something, some actual act, meaningful act has been done by the uttering of the words, I thank you for this, or I offer my congratulations or something like that.

If someone doesn't do that, you know, then you infer all sorts of things and there's some significance to it. Declarations are, I think, the canonical example of speech acts. This is when saying the thing actually does a thing. So, ceremonies are a good example of this, and the example that's always used is marriage. So if someone says, you know, I pronounce you, husband and wife, that's not just a thing that they said. Saying it actually does it. The saying of the phrase actually carries out the action. The action of making them of wedding the people.

Another example is pronouncing someone guilty. If a judge says, I find you guilty, that action is the act of making them or finding them guilty. It's not a statement that it happens somewhere else. It's the doing of the action itself. Religious rituals or other things like that, or if you were swirming someone into an office or something like that, all these sorts of things would be examples of declarations

in which you sort of change reality in some way in accordance with the proposition that's being stated. You know, I repudiate this contract would be a speech act which would make, be changing reality in a way such that it accords with what you've just said. It's not just, you're not just saying something, you're doing something by the saying of it. Or, I declare war, that might be another example, if you're in some position of authority. You're changing something about the world by making that declaration. It's an act. It's not just a thing that you say.

So speech act theory, I think, is very interesting and can offer lots of interesting. insights about how language and culture and society interact. So, that brings us to a conclusion of this series of episodes. Hopefully you've found it interesting. Jump onto the Facebook page. If you go to Facebook and just search for the Science of Everything podcast, you should be able to find it. And there you'll be able to find links to past episodes and also a lot of visual material

which I put up to help support the stuff that I'm talking about in the podcast. I hope you understand it a bit more. Also, I love to hear from listeners. Send me an email at F-O-D-S-1. 1.2 at gmail.com, Fods12 at gmail.com. Also, I'd be very grateful if you go to iTunes and post a favorable review or even just a star rating. That really helps to keep the visibility of the podcast up, and so I much appreciate it if you do that. So, hopefully you've found this interesting.

Thanks for listening, and I'll talk to you next time.